macedonizer commited on
Commit
bb96823
1 Parent(s): b714afc

initial commit

Browse files
README.md ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - mk
4
+ thumbnail: https://huggingface.co/macedonizer/mk-roberta-base/blaze-koneski.jpg
5
+ license: Apache 2.0
6
+ datasets:
7
+ - wiki-mk
8
+ - time-mk-news-2010-2015
9
+ ---
10
+
11
+ # mk-gpt2
12
+ Test the whole generation capabilities here: https://transformer.huggingface.co/doc/gpt2-large
13
+ Pretrained model on English language using a causal language modeling (CLM) objective. It was introduced in
14
+ [this paper](https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf)
15
+ and first released at [this page](https://openai.com/blog/better-language-models/).
16
+
17
+ ## Model description
18
+ mk-gpt2 is a transformers model pretrained on a very large corpus of Macedonian data in a self-supervised fashion. This
19
+ means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots
20
+ of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely,
21
+ it was trained to guess the next word in sentences.
22
+ More precisely, inputs are sequences of continuous text of a certain length and the targets are the same sequence,
23
+ shifted one token (word or piece of word) to the right. The model uses internally a mask-mechanism to make sure the
24
+ predictions for the token `i` only uses the inputs from `1` to `i` but not the future tokens.
25
+ This way, the model learns an inner representation of the Macedonian language that can then be used to extract features
26
+ useful for downstream tasks. The model is best at what it was pretrained for however, which is generating texts from a
27
+ prompt.
28
+
29
+ ### How to use
30
+ Here is how to use this model to get the features of a given text in PyTorch:
31
+
32
+ import random
33
+ from transformers import AutoTokenizer, AutoModelWithLMHead
34
+
35
+ tokenizer = AutoTokenizer.from_pretrained('macedonizer/mk-gpt2') \
36
+ model = AutoModelWithLMHead.from_pretrained('macedonizer/mk-gpt2')
37
+
38
+ input_text = 'Скопје е '
39
+
40
+ if len(input_text) == 0: \
41
+ encoded_input = tokenizer(input_text, return_tensors="pt") \
42
+ output = model.generate( \
43
+ bos_token_id=random.randint(1, 50000), \
44
+ do_sample=True, \
45
+ top_k=50, \
46
+ max_length=1024, \
47
+ top_p=0.95, \
48
+ num_return_sequences=1, \
49
+ ) \
50
+ else: \
51
+ encoded_input = tokenizer(input_text, return_tensors="pt") \
52
+ output = model.generate( \
53
+ **encoded_input, \
54
+ bos_token_id=random.randint(1, 50000), \
55
+ do_sample=True, \
56
+ top_k=50, \
57
+ max_length=1024, \
58
+ top_p=0.95, \
59
+ num_return_sequences=1, \
60
+ )
61
+
62
+ decoded_output = [] \
63
+ for sample in output: \
64
+ decoded_output.append(tokenizer.decode(sample, skip_special_tokens=True))
65
+
66
+ print(decoded_output)
added_tokens.json ADDED
@@ -0,0 +1 @@
 
1
+ {"<|endoftext|>": 52000, "<|beginoftext|>": 52001, "<PAD>": 52002}
blaze-koneski.jpg ADDED
Binary file
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "../../models/mk-gpt2",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2LMHeadModel"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 0,
9
+ "embd_pdrop": 0.1,
10
+ "eos_token_id": 2,
11
+ "gradient_checkpointing": false,
12
+ "initializer_range": 0.02,
13
+ "layer_norm_epsilon": 1e-05,
14
+ "model_type": "gpt2",
15
+ "n_ctx": 1024,
16
+ "n_embd": 768,
17
+ "n_head": 12,
18
+ "n_inner": null,
19
+ "n_layer": 12,
20
+ "n_positions": 1024,
21
+ "resid_pdrop": 0.1,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.9.1",
30
+ "use_cache": true,
31
+ "vocab_size": 52003
32
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3a81e38ee1b4a5a90f3e229bff5c81483e78269919e8b4f1e1d96847e370667
3
+ size 515767529
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
1
+ {"bos_token": "<|beginoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<unk>", "pad_token": "<PAD>", "mask_token": "<mask>"}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
1
+ {"errors": "replace", "unk_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "special_tokens_map_file": null, "name_or_path": "../../models/mk-gpt2", "tokenizer_file": "../../models/mk-gpt2\\tokenizer.json", "tokenizer_class": "GPT2Tokenizer"}
vocab.json ADDED
The diff for this file is too large to render. See raw diff